All notes

AI

May 7, 2026

OpenAI President Defends Personal Diary Entries Under Oath in Ongoing Trial

Greg Brockman reads personal diary entries aloud in court as part of legal proceedings examining OpenAI's conduct and internal motivations around its nonprofit-to-for-profit transition.

Greg Brockman, president of OpenAI, faced an unusual moment in court: reading his own personal diary entries aloud to a jury. The entries, described as sounding greedy, became evidence in legal proceedings scrutinizing OpenAI's internal culture and decision-making.

The trial context matters. OpenAI has been navigating sustained legal and regulatory pressure over its structural shift away from a nonprofit governance model. Elon Musk's lawsuit against the company and its founders centers on allegations that the transition prioritizes profit over the original mission. The diary entries surfaced in that context suggest internal documentation is now fair game for discovery.

For engineers and technical founders watching this space, the implication is straightforward: private communications, journals, and internal notes at AI companies are no longer safely informal. When organizations operate at the scale and influence OpenAI does, the boundary between personal reflection and discoverable evidence collapses under litigation pressure.

Brockman's position as president makes his personal writings particularly relevant to arguments about intent and organizational culture. Plaintiffs seeking to establish that leadership acted in self-interest rather than in accordance with the nonprofit charter will use any contemporaneous documentation available. A diary written in the moment carries more weight than testimony constructed after the fact.

The broader signal for the AI industry is procedural. Founders and executives at well-capitalized AI startups should treat personal notes about business decisions with the same care as formal communications. Discovery in high-stakes litigation is broad. Anything written during a period of organizational controversy can be subpoenaed.

This trial is unlikely to resolve quickly. The structural questions about OpenAI's governance remain contested, and testimony like Brockman's adds texture to what is fundamentally a dispute about fiduciary duty and mission fidelity at one of the most consequential AI labs operating today.